This project represents a new paradigm for epidemiologic research on alcoholism: it nests a high risk genetic/family study within an epidemiologic framework. Advantages over existing studies are: ascertainment of families from a representative sample; systemic selection of families at high and low risk of alcoholism; direct assessments of multiple family members, including up to 3 offspring and parents; longitudinal follow-up of the offspring; over-sample of minority families; and collection of DNA for future assessments of genetic risk/protective factors. This design permits testing of hypothesis relating to sibling and peer influences on adolescent/young adult alcohol problems, which more conventional parents-one child designs, and even more specialized twin-family designs, may tested in a limited way. Families are ascertained through birth records of adolescents and young adults born in Missouri (ensuring comparability with Project 1), and biological mothers administered a telephone screening interview to identify potentially high risk families. All families screening positive, and a subsample of those screening negative will be invited for the second stage, where biological mothers and fathers (or best informant) will be administered a telephone diagnostic interview and question to assess own and co-parent's history of DSM-I alcohol abuse or dependent, as well as history of behavioral and emotional problems of the proband and up to 2 siblings aged 13-25. Probands and up to 2 of their siblings from families where fathers were reported by mothers to have 3 or more symptoms of alcohol dependence ("high risk", 100 AA, 100 W) and those from families where fathers were not so reported, or who screened negative ("controls": 150 AA, 150 W) will be invited for telephone diagnostic interview at baseline and again at 2 years, with brief telephone diagnostic interview at baseline and again at 2 years, with brief telephone follow up interviews in other years along with annual questionnaire assessments. Data will be used to test hypotheses about (i) sibling and peer influences on alcohol and other substance involvement and problems, (ii) relationships between parenting style, parental alcoholism and co-morbidity, and offspring alcoholism, (iii) relationships between co-morbidity, temperament, traumatic events and this chronic stress, and offspring alcoholism, and (iv) the modification of these relationships by ethnicity. Results from this project will provide an important check on the generalizability of finds from other projects that rely upon specialized twin-family designs.